Safeguarded Anderson acceleration for parametric nonexpansive operators
This paper describes the design of a safeguarding scheme for Anderson acceleration to improve its practical performance and stability when used for first-order optimisation methods. We show how the combination of a nonexpansiveness condition, conditioning constraints, and memory restarts integrate w...
| Main Authors: | , , |
|---|---|
| Format: | Conference item |
| Language: | English |
| Published: |
IEEE
2022
|
| Summary: | This paper describes the design of a safeguarding
scheme for Anderson acceleration to improve its practical
performance and stability when used for first-order optimisation methods. We show how the combination of a nonexpansiveness condition, conditioning constraints, and memory
restarts integrate well with solver algorithms that can be
represented as fixed point operators with dynamically varying
parameters. The performance of the scheme is demonstrated
on seven different QP and SDP problem types, including more
than 500 problems. The safeguarded Anderson acceleration
scheme proposed in this paper is implemented in the opensource ADMM-based conic solver COSMO. |
|---|